{"skill":{"slug":"agent-mail-guard","displayName":"Agent Mail Guard — Email Sanitizer for AI Agents","summary":"Sanitize email and calendar content before it reaches your AI agent's context window. Blocks prompt injection, markdown image exfiltration, invisible unicode...","tags":{"email":"1.4.0","latest":"1.4.0","openclaw":"1.4.0","prompt-injection":"1.4.0","sanitizer":"1.4.0","security":"1.4.0"},"stats":{"comments":0,"downloads":371,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1772229832892,"updatedAt":1777525453581},"latestVersion":{"version":"1.4.0","createdAt":1772229832892,"changelog":"- Enhanced description and documentation for clarity and practical usage.\n- Outlines attack vectors detected and the corresponding actions taken.\n- Details the output JSON structure, sender trust tiers, and specific agent integration safety rules.\n- Improved customization guidance, including trust contacts and detection patterns.\n- Added architecture and local processing information.\n- Included testing instructions and requirements.","license":null},"metadata":{"os":null,"systems":null},"owner":{"handle":"discodaddy","userId":"s1782hxffnae9p2w1b85ac1d2d84fb4c","displayName":"DiscoDaddy","image":"https://avatars.githubusercontent.com/u/130884525?v=4"},"moderation":{"isSuspicious":true,"isMalwareBlocked":false,"verdict":"suspicious","reasonCodes":["suspicious.prompt_injection_instructions"],"summary":"Detected: suspicious.prompt_injection_instructions","engineVersion":"v2.4.5","updatedAt":1777525453581}}